Building AI models that understand chemical principles
AI Summary: MIT Associate Professor Connor Coley is advancing the use of artificial intelligence in small-molecule drug discovery by developing computational models that analyze and predict chemical compounds and reaction pathways. His research integrates machine learning and cheminformatics to optimize automated chemical reactions, facilitating the identification of potential drug candidates from an estimated 10^20 to 10^60 possible compounds. Coley's work, supported by DARPA's Make-It program, aims to enhance the synthesis of medicines and other useful compounds, demonstrating the intersection of AI and chemical engineering. His academic journey includes a postdoctoral position at the Broad Institute, where he focused on identifying small molecules from extensive candidate libraries.